Reappraisal of lymphatic mapping for

midgut neuroendocrine patients

undergoing cytoreductive surgery
Yi-Zarn Wang, DDS, MD, Jean P. Carrasquillo, MD, Elizabeth McCord, BS, Rhea Vidrine, BS,
Monica L. Lobo, BS, S. Ali Zamin, BS, Philip Boudreaux, MD, and Eugene Woltering, MD,
New Orleans, LA

Background. We previously reported that midgut neuroendocrine tumors (NETs) often develop
alternative lymphatic drainage owing to lymphatic obstructions from extensive mesenteric lymphade-
nopathy, making intraoperative lymphatic mapping mandatory. We hypothesize that this innovative
approach needs a longer term validation.
Methods. We updated our results by reviewing 303 patients who underwent cytoreduction from
November 2006 to October 2011. Of these patients, 112 had lymphatic mappings and 98 were for
midgut NET primaries. Among them, 77 mappings were for the initial cytoreduction and 35 were for
reexploration and further cytoreduction. The operative findings, pathology reports, and long-term
surgical outcomes were reviewed.
Results. Lymphatic mapping changed traditional resection margins in 92% of patients. Of the 35
patients who underwent reexploration without initial mapping, 19 (54%) showed a recurrence at or
near the anastomotic sites. In contrast, none of the 112 mapped patients had shown signs of recurrence
in a 1- to 5-year follow-up. Additionally, 20 of 45 ileocecal valves (44.4%) were spared in patients
whose tumors were at the terminal ileum that, traditionally, would call for a right hemicolectomy.
Conclusion. With a longer follow-up, lymphatic mapping has proven to be a safe and effective way to prevent
local recurrences and preserve the ileocecal valve for selected patients. (Surgery 2014;156:1498-503.)

From the Division of Surgical Oncology, Department of Surgery, Louisiana State University Health Sciences
Center–New Orleans, New Orleans, LA

MIDGUT NEUROENDOCRINE TUMORS (NETs) are rare ma- subsequent drop metastases, and small bowel
lignancies with an indolent course. The diagnosis is infested with potentially microtumor emboli can
often delayed until advanced stages of the disease easily be missed by the surgeon’s and become sites
have been reached.1 The only reliable and durable of missing residual tumors or recurrences after
treatment of the primary tumors and local regional resection. Thus, intraoperative lymphatic mapping
lymphadenopathy is surgical cytoreduction.2-5 becomes essential and mandatory.6
We previously reported that midgut NETs often We hypothesized previously that traditional
develop alternative lymphatic drainage owing to resection margins may be inadequate and likely
lymphatic obstructions from extensive mesenteric increase local recurrence at the anastomosis. We
lymphadenopathy. These alternative subserosal also argued that a traditional right hemicolectomy
lymphatic drainages often lead to multiple drop for midgut NET patients whose tumors were located
metastases that in the past were erroneously close to the cecum may be overly aggressive and can
interpreted as ‘‘multiple primaries’’ in #30% of lead to an even poorer quality of life with worsening
patients who presented with midgut NET. These diarrhea. We deem formal right hemicolectomies
extended alternative lymphatic drainage pathways, unnecessary when the lymphatic mapping clearly
demonstrates sparing of lymphatic drainage from
Accepted for publication May 27, 2014.
the terminal ileum tumor into the cecum or cecal
Reprint requests: Elizabeth McCord, BS, Louisiana State Univer-
mesentery. We recommended lymphatic mapping
sity Health Sciences Center–New Orleans, School of Medicine, to define the resection margins and to preserve
Secretary, Class of 2015, New Orleans, LA 70123. E-mail: ileocecal valve in selective patients. The results of
[email protected]. our initial experience were reported in 2009.
0039-6060/$ - see front matter We then further hypothesized that this innova-
Ó 2014 Elsevier Inc. All rights reserved. tive approach needs a longer term validation to
http://dx.doi.org/10.1016/j.surg.2014.05.028 prove that traditional ‘‘eyeball’’ resection margins

1498 SURGERY
Surgery Wang et al 1499
Volume 156, Number 6

are grossly inadequate and carry an increased

likelihood of local recurrence. In addition,
lymphatic mapping may safely preserve the ileoce-
cal valve for selected patients without late recur-
rence in the retained cecum or ascending colon.

METHODS
We updated our results by reviewing 605 NETs
patients in our database. From November 2006 to
October 2011, 303 patients underwent cytoreduc-
tion. Of these patients, 112 had lymphatic map-
pings for gastrointestinal NET primaries and 98
were for midgut NET primaries. Seventy-seven
lymphatic mappings were performed for the initial
cytoreduction and 35 were mapped in reexplora-
Fig 1. Blue dye injected at the primary, most proximal,
tion for further cytoreduction. Demographically, and most distal sites of multiple tumors.
there are 59 males and 43 females included in the
study with a mean age of 59.4 years (range, 20–79).
The majority of the study’s patients (77.5%)
presented with stage IV disease at the time of
cytoreduction/mapping; the rest of the cohort
(22.5%) presented with stage III disease. The
average operative time was 376 ± 129 minutes.
The mean follow-up time for the entire cohort was
24.4 months (range, 1–60). All patients included
in the study were followed closely with a standard
protocol established in our clinic. This protocol
consists of collecting tumor markers every
3 months, restaging image(s) to establish a new
baseline 3–6 months postoperatively, and imaging
studies 6 months thereafter as clinically indicated
or per physicians’ discretions.
All lymphatic mappings were performed in a
standard fashion. Once the midgut lesion identi-
fied, 2–5 mL of lymphazurin blue dye (Covidien,
Norwalk, CT) was injected peritumorly around the
single primary tumor into the subserosal region of
the small bowel. In cases of multiple tumors, blue Fig 2. Resection of small bowel mesentery following
dye was injected around the most proximal, the blue dye guidance.
most distal, and the primary tumors respectively
(Fig 1). The small bowel was then tucked away to Otherwise, a segmental small bowel resection was
allow the mapping to be completed spontaneously. performed without a formal right hemicolectomy
While giving the dye 10–15 minutes to migrate to retain the ileocecal valve. The intraoperative find-
through the lymphatics, a cholecystectomy, mobili- ings and pathology reports were then reviewed. Re-
zation of the liver, and/or intraoperative liver ultra- sected margins defined by lymphatic mapping were
sonography was performed. Once the lymphatic compared with traditional surgical margins of 5 cm
mapping was complete, resection of the small bowel proximal and distally. The incidence of ileocecal
segment containing the NETs and the adjacent mes- valve preservation was investigated. The long-term
entery was then conducted following blue dye guid- surgical outcomes were also reviewed to evaluate
ance (Fig 2). All tissues stained by the blue dye are the safety and efficacy of mapping and ileocecal
removed including the mesenteric lymph nodes in valve preservation based on mapping.
otherwise unsuspected regions (Fig 3). For tumors
located close to the ileocecal valve, a hemicolectomy RESULTS
was only performed for patients whose blue dye had No adverse events were observed during the
traversed to the cecum or colonic mesentery (Fig 4). 112 lymphatic mapping procedures. Lymphatic
1500 Wang et al Surgery
December 2014

lab testing and imaging studies in a 1- to 5-year

follow-up among the 112 mapped patients. Addi-
tionally, 20 of the 45 ileocecal valves (44.4%) were
spared in patients whose tumors were at the
terminal ileum based on mapping that demon-
strated clearance of cecum and or cecal mesentery
involvement. Traditionally, these patients would
have required a right hemicolectomy, which can
worsen diarrhea for carcinoid patients. More
important, not 1 of the 20 patients who underwent
an ileocecal valve sparing operation has displayed
recurrent disease in their right colon or right
colon mesentery thus far.
The primary goal of our cytoreductive surgeries
with lymphatic mapping for midgut NET patients
is to remove the small bowel primary, drop
metastasis in adjacent bowel segments, and
regional lymphadenopathy in an attempt to
achieve R0 resection. While allowing the dye to
migrate, a cholecystectomy is performed to ensure
patients can endure long-term somatostatin analog
treatments without the future threat of cholecys-
Fig 3. Mesenteric lymph nodes in an otherwise unsus-
titis. All patients undergo maximal cytoreductive
pected region.
surgery as intraoperative conditions permit to
commence the second phase of our operation. If
the patient remains stable after resection of the
primary and nodes, the debulking is then
extended to remove metastasis of liver, pancreas,
diaphragm, pelvis, perineum, and occasionally
crossing the diaphragm to enucleate pulmonary
metastasis. We recommend and perform routinely
oophorectomies for perimenopausal woman
because our previous study had indicated that
18% of female patients would suffer from such
metastasis during their disease’s progression (un-
published data). Worst of all, such metastases are
associated with increasing incidence of right-sided
heart valvular disease and carcinoid syndromes
owing to the fact that gonadal venous drainage
Fig 4. Terminal ileum tumor with blue dye traversing bypasses hepatic clearance.
the cecum and colonic mesentery.
DISCUSSION
Midgut NETs are rare malignancies with an
mapping changed traditional resection margins in incidence of 5.25 per 100,000 individuals.7 This
92% of patients with 100% negative resection low-grade malignancy typically presents with vague
margins on final pathology confirmation. Most symptoms and an indolent course. The diagnosis is
intriguingly, of the 35 patients who underwent often delayed until advanced stages of the disease
reexploration without mapping at their first oper- have been reached.1 It has been reported that
ation, 19 patients (54%) were found to have a 30% of midgut NET patients have multiple pri-
recurrence at or near their original anastomotic maries and a high propensity to recur after surgical
sites. This finding sustains our previous hypothesis intervention. As a result, a passive treatment
that traditional margin without lymphatic mapping course for midgut NET patients has been formu-
are grossly inadequate. In contrast, with our lated and implemented for years. Only recently,
vigorous clinical surveillance, no recurrence has surgeons from a few medical centers with special
been detected with currently available diagnostic interest in NETs began to report promising and
Surgery Wang et al 1501
Volume 156, Number 6

durable outcomes with surgical resection of pri-

mary tumors and debulking of regional and distant
disease even for stage IV patients.2-5
While striving to reduce postsurgical local
recurrence and improving the overall quality of
lives of midgut NET patients, we began applying
intraoperative lymphatic mapping to define resec-
tion margins in late 2006. In addition, we utilize
the mapping to spare ileocecal valves in patients
whose lymphatic drainage clearly bypasses the
cecum and or cecal mesentery. In 2009, we first
published our series of 49 patients who underwent
lymphatic mapping for midgut carcinoids. This
technique for defining resecting margins was pro-
posed based on the observation that the bulky
Fig 5. Longitudinal lymphatic drainage along subserosal
mesenteric adenopathy often seen in these pa- lymphatic channels.
tients with longitudinal lymphatic drainage along
subserosal lymphatic channels (Fig 5) who also
typically present with multiple ‘‘primaries’’
(Fig 6). We hypothesize that alternative lymphatics
lead to multiple drop metastases, microscopic
tumor emboli, or seeding in adjacent, grossly
normal-appearing bowel, which explains the high
local recurrence rates after traditional surgery
without mapping. We also dispute the notion
that 30% of midgut NET patients have multiple
primaries. In fact, these tumors are in reality
drop metastases from a singular primary based
on a genetic analysis.8 With mapping, we believe
surgeons can improve their cytoreductions and
decrease local recurrence of NETs by removing tu-
mor burden (grossly visible and/or microscopic
drop metastasis) and any potential tumor-bearing
lymphatics, concomitantly. We also promote ileoce-
cal valve preservation in appropriate candidates to
improve the quality of life for patients with midgut
Fig 6. Multiple ‘‘primaries.’’
NETs who have suffered from serotonin over surge
and long-term bowel ischemia.
We understand that our first report had a
notably brief follow-up period and required a It is difficult to prove the direct survival advan-
longer follow-up time to validate our proposal. tages of implementing surgical modifications,
For 8 years, we have continued our practice of namely, lymphatic mapping and sparing ileocecal
mapping and have acquired data for long-term valves for the cytoreduction of an indolent disease
follow-up. Now, with this updated reappraisal in such as midgut NETs. Only a long-term, close
hand, we would like to reconfirm our hypothesis follow-up study will be able to dismiss or sustain
that mapping is a time-efficient and safe procedure such a claim. The current reappraisal study was
that improves the degree of cytoreduction, negates conducted for this purpose. We plan to continue
arbitrary margins, preserves intestinal length when our use of lymphatic mapping while perpetually
possible, and yet removes all potential cancer collecting data to monitor the long-term effects of
infested bowel while sparing the ileocecal valve in this innovative procedure. Nonetheless, our in-
selected patients. With all of its merits, surgeons stitution’s recently reported data have shown that
can expect mapping to help reduce local recur- aggressive cytoreduction surgery with the applica-
rence, lessen postoperative diarrhea, and improve tion of lymphatic mapping and a multidisciplinary
quality of life, which may eventually translate into approach has dramatically improved the tradition-
better long-term survival. ally reported 5- and 10-year survival rates of midgut
1502 Wang et al Surgery
December 2014

NETs patients from 54% and 30% to 87% and Dr Jim Howe (Iowa City, IA): One question I
77%, respectively.9 Perhaps these much improved have is since most people have 3–5 m of small
survival figures can be used tangentially to support bowel, why not just take wider margins? Do we
the use of lymphatic mapping in the surgical the- really need to do the lymphatic mapping? Sec-
ater for midgut NETs patients undergoing cytore- ond, how do you feel about preserving the ileoce-
ductive surgery. However, a much protracted cal valve when you have taken the ileocolic
study looking at the surgical outcomes of mapped artery?
versus unmapped patients will provide us with a Ms Elizabeth McCord: I will address your sec-
more precise answer. ond question regarding the preservation of the
ileocecal valve first: 73% of our carcinoid patients
have diarrhea at their initial presentation to our
REFERENCES
clinic. In knowing the morphology and function
1. Modlin IM, Lye KD, Kidd M. A 5-decade analysis of 13,715
of the valve, we understand that sparing the valve
carcinoid tumors. Cancer 2003;97:934-59.
2. Ohrvall U, Eriksson B, Juhlin C, Karacagil S, Rastad J, Hell- should improve their quality of life by minimizing
man P, et al. Method for dissection of mesenteric metastases their diarrhea. As for the colic artery, I wish I had
in mid-gut carcinoid tumours. World J Surg 2000;24:1402-8. a better answer for your question. That is some-
3. Soreide JA, van Heerden JA, Thompson GB, et al. Gastroin- thing I will definitely consider in my future
testinal carcinoid tumors: long-term prognosis for surgically
research. Dr Wang would like to address this
treated patients. World J Surg 2000;24:1431-6.
4. Hellman P, Lundstr€ €
om T, Ohrvall U, et al. Effect of surgery question.
on the outcome of midgut carcinoid disease with lymph Dr Yizarn Wang (New Orleans, LA): We do not
node and liver metastases. World J Surg 2002;26:991-7. have any problem with preserving the ileocecal
5. Gulec SA, Mountcastle TS, Frey D, et al. Cytoreductive valve so far. I think it is because, most of the
surgery in patients with advanced-stage carcinoid tumors.
time, we did not disrupt the middle colic. Also,
Am Surg 2002;68:667-71.
6. Wang Y-Z, Joseph S, Lindholm E, Lyons J, Boudreaux JP, the right colic is still there, so even if we take the
Woltering EA. Lymphatic mapping helps to define resection ileocolic vessel, the cecum survives without
margins for midgut carcinoids. Surgery 2009;146:993-7. problems.
7. Yao JC, Hassan M, Phan A, et al. One hundred years after Dr Mary Beth Hughes (Bethesda, MD): You are
‘‘carcinoid’’: epidemiology of and prognostic factors for
a tertiary referral center for these patients and,
neuroendocrine tumors in 35,825 cases in the United States.
J Clin Oncol 2008;26:3063-72. based on your data, it looks like they are later stage
8. Guo Z, Li Q, Wilander E, Ponten J. Clonality analysis of patients. I know that it changed your surgical resec-
multifocal carcinoid tumors of the small intestine by tion in 92% of cases, but, in earlier stage patients,
X-chromosome inactivation analysis. J Pathol 2000;190:76-9. where they just have smaller positive nodes, it is
9. Boudreaux JP, Wang YZ, Diebold AE, Frey DJ, Anthony L,
not clear to me that the drainage is altered
Uhlhorn AP, et al. A single institution’s experience with
surgical cytoreduction of stage-IV, well-differentiated, small much. Can you talk about your experience with
bowel neuroendocrine tumors (NETs). J Am Coll Surg lower stage patients, where they had microscopi-
2014;218:837-44. cally positive nodes in were stage III, and not the
big bulky, trying to eat the superior mesenteric
DISCUSSION artery and nodes?
Dr Richard Hodin (Boston, MA): The assump- Ms Elizabeth McCord: Unfortunately, our de-
tion is that the lymphatic mapping probably makes mographics for this study are almost exclusively
you take more, not less. I guess I am wondering stage III and IV patients. In our study, we looked
whether you have actual data on that, the amount at approximately 77% of patients with stage IV dis-
of bowel resected sort of using lymphatic mapping ease, so they already have metastasis to the liver
versus not. and boggy mesentery lymphadenopathy. In terms
Ms Elizabeth McCord (New Orleans, LA): In of the drainage pattern in patients with only micro-
our results, we showed that we are actually modi- scopic positive lymph nodes, I do not have an
fying the length of bowel resections by 92% of answer for you at this time. Dr Wang would like
the time. I do not have the exact centimeters to to address your question.
which we extended the resections. On average, Dr Yizarn Wang: The great majority of our
our resection margin is about 7–10 cm with the patients are coming on a referral basis. We have
longest margin of #27 cm. This is much greater patients from 46 different states and 6 different
than the traditional margins of 5 cm. We have countries; 48% have had their primary removed
taken #120 cm of bowel in some patients. The already by the time they were referred to me.
overwhelming majority do have more bowel They were sent to us because of lymphadenopathy
resected. encasing the mesenteric vessels making them
Surgery Wang et al 1503
Volume 156, Number 6

difficult surgical candidates. We often take respon- each patient before 2011. When we were reviewing
sibility of further care for those patients. With our data, we noted only about one third of patients
some of our remapped patients, we did find a re- had been mapped. Since 2011, our group has
sidual tumor, making us believers in mapping. routinely mapped every single patient on whom
Initially, not every surgeon in our group mapped we operated.